How to set a media budget when the board wants growth and finance wants certainty
Media budgets fail when they are set as a single number. Set them as a portfolio with different jobs and different standards of proof.
The annual budget conversation usually goes wrong the same way. The board wants ambitious growth, finance wants predictable returns, and marketing is asked to promise both with one number. The way out is to stop treating media spend as one thing.
A budget structured as a portfolio, with three layers that each have a different job, survives contact with a bad month far better.
The core, the growth bets, and the experiments
The core is spend with proven, measured returns: brand search, high-intent non-brand, retargeting with honest frequency caps. It carries the strictest efficiency targets and defends the revenue base.
Growth bets are scaled expansions of things that already worked at smaller size: new markets, new campaign types, demand creation on YouTube or social. They run on leading indicators and checkpoint reviews rather than immediate ROAS.
Experiments are small, capped allocations for genuinely new ideas, judged on how fast they teach you something. Most will fail, which is exactly why they stay small.
Agree the rules before the money moves
The proportions matter less than the agreement. Seventy percent core, twenty growth, ten experiments is a reasonable starting point, adjusted for how mature the account is.
What finance gets is certainty about the core. What the board gets is a real growth mechanism instead of a hope. And what marketing gets is the ability to lose a test without losing the argument, because everyone already agreed what each pound was for.